Article Text

Download PDFPDF

How to diagnose and treat aerobic and desquamative inflammatory vaginitis
  1. Mark J Mason,
  2. Andrew J Winter
  1. Sandyford Sexual Health Services, Glasgow, UK
  1. Correspondence to Mark J Mason, Sandyford Sexual Health Services, 2-6 Sandyford Place, Glasgow G3 7NB, UK; Mark.mason{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Abnormal vaginal discharge is a common problem which can usually be managed syndromically in non-specialist services.1 Common pathological causes include bacterial vaginosis (BV), candidiasis and Trichomoniasis (TV). Some women fail to improve with syndromic treatment, or present with additional symptoms such as severe dyspareunia or purulent discharge. This article focuses on a practical approach to investigating and managing such a presentation in a specialist setting where near-patient microscopy is available. It needs to be read in conjunction with guidelines, such as the 2011 European (IUSTI/WHO) Guideline on Management of Vaginal Discharge.2

Overgrowth conditions of the vagina

Normal vaginal health in adult women is maintained by acid-secreting lactobacilli. In overgrowth conditions such as BV these healthy lactobacilli are disrupted, although the underlying cause for this remains unclear. In all these conditions vaginal pH rises and there is usually an increased discharge.

BV presents with a painless but unpleasant, odorous discharge. There is overgrowth of anaerobic organisms, such as Gardnerella vaginalis, Mobiluncus spp., BV-associated bacteria 1–3 and Atopobium spp. causing a rise in vaginal pH, discharge, and smell, but little inflammation. Once BV is established, biofilm formation may explain the high recurrence rate.3

Aerobic vaginitis (AV) presents with a more purulent discharge with significant inflammation and epithelial disruption. Patients may also report burning, stinging and dyspareunia. There is predominance of aerobic flora such as Escherichia coli, group B streptococci, Staphylococcus aureus and evidence of local cytokine-mediated inflammation.4

Desquamative inflammatory vaginitis (DIV), is a chronic condition of unknown aetiology with vaginal rash and purulent discharge. Patients present with profuse discharge, vestibulovaginal irritation, dyspareunia and vaginal inflammation or erythema. Many patients have long-standing symptoms for 12 months or more. Vaginal flora is altered, probably secondary to the underlying epithelial damage.5


Clinical approach

Take a detailed clinical and sexual history, including record of previous failed treatments, both prescribed and self-taken, and a menstrual history. During speculum examination note the discharge colour, consistency, amount and odour, as well as any atrophic change. Record vaginal pH from the vaginal wall using suitable narrow-range pH paper (eg, Whatman, pH 4–7). Obtain material from the lateral vaginal walls with a sterile 10 µl plastic loop or swab and spread evenly onto two clean glass microscope slides. Keep one for Gram staining, and then make a wet preparation by adding a small drop of sterile physiological saline (0.9%) and adding a coverslip. A high vaginal swab (HVS) for culture should be taken if the patient has recurrent discharge, or is pregnant, post partum, post abortion or post cervical instrumentation. A routine sexual health check may be indicated by the history which should include tests for Chlamydia, Gonorrhoea, HIV and Syphilis. Consider other causes of vaginal discharge, such as retained foreign body, ectopy, fistula, malignancy, dermatoses such as erosive lichen planus and allergic reactions.

Microscopy: Gram stain for modified Hay-Ison score

A Gram-stained slide of vaginal discharge should be examined under ×1000 magnification and scored using modified Hay-Ison criteria.6 Investigations are best carried out in a setting where on-site microscopy is available by trained personnel and specimens can be freshly examined. Clinical staff undertaking microscopy can access a training course run by the British Association for Sexual Health and HIV (see On-site microscopy should be subject to a relevant laboratory accreditation programme that meets ISO 15189:2012, such as that overseen by the UK Accreditation Service (

Microscopy: wet film for AV score

In more challenging cases we have found it valuable to apply a more complex scoring algorithm, such as the ‘AV’ score published by Donders in 2002.4 The AV score combines information about bacterial flora, epithelial disruption and inflammation to yield a value between 0 and 10. We have slightly simplified the original definitions in a short scoring sheet (figure 1). In our hands this takes around 1–3 min to perform.

Figure 1

Abbreviated template for assessing the aerobic vaginitis score (adapted from Donders4).

To assess the AV score, examine a wet film of posterior fornix fluid at ×400 by phase contrast. Examine:

  1. the background flora: are there coliforms or lots of cocci or chained bacteria? (see figure 2).

  2. the lactobacillary grade: are lactobacilli present in normal numbers, reduced or absent?

  3. leucocyte numbers: number and proportion of these compared with epithelial cells.

  4. proportion of ‘toxic’ leucocytes: these exhibit lysozymal activity, which can sometime be better seen under ×1000 dark ground where facilities and time permit (see online supplementary file-video).

  5. proportion of parabasal epithelial cells: immature round cells, with a refractile border and a large dense nucleus (see figure 3).

Figure 2

Wet film preparation of vaginal discharge examined under phase contrast at ×400 magnification, showing lactobacillary grade III (with chained cocci, arrow), aerobic vaginitis score contribution 2.

Figure 3

Wet film preparation of vaginal discharge examined under phase contrast at ×400 magnification showing numerous parabasal cells (asterisk) and numerous leucocytes (arrow). The parabasal cells suggest significant epithelial disruption. Aerobic vaginitis score was 6 overall.

The Gram stain can be useful for confirming any abnormal flora seen on the AV score as it uses the higher magnification of ×1000.

We notify clinicians of AV scores of 5 or over, as this represents significant disturbance of bacterial flora combined with significant inflammation or epithelial damage. Lower scores are non-specific: for example, straightforward BV will score as AV=2, due to loss of lactobacilli.

Microbiological tests

Cultures prepared from an HVS can exclude Candida spp., anaerobes, Streptococcus pyogenes and staphylococcal infection. Some labs may suppress reporting of what they consider commensal organisms, so in a patient where AV is suspected, ask the microbiologist to report all organisms identified on culture. Rapid tests or nucleic acid amplification test (NAAT) for TV offer better sensitivity than culture to exclude trichomoniasis. In future, real-time PCR techniques7 or next-generation sequencing might shed more light on disturbances in the vaginal microbiome.8


In our own unit around 5% of women presenting with vaginal discharge have AV scores of 5 and over. Many will have modified Hay-Ison scores consistent with BV. In this small subgroup we believe a moderate or severe AV score should trigger careful review of other causes of vaginal discharge and discomfort before managing for BV alone.

Steps include:

  • Assess degree of atrophic change, remembering this can occur with long-acting progesterone depot contraception as well as early or natural menopause.

  • Careful retesting for trichomoniasis, although less needed with the increasing first line use of more sensitive near-patient and molecular amplification tests.

  • Direct plating of vaginal swab on to Sabouraud agar which may increase sensitivity for candida isolation.

  • Examine for other signs of lichen planus and lichen sclerosus.

  • Consider other symptoms and illness in close contacts that may indicate a Group A streptococcal infection.

Women with significantly abnormal AV scores are very heterogenous, which makes interpreting the few treatment studies published very difficult. In general authors prefer antimicrobials active against aerobic bacteria such as topical clindamycin,4 kanamycin ovules9 or moxifloxacin.10 Atrophy needs treatment with oestrogen replacement.11 Literature on DIV suggests combination use of intravaginal clindamycin and intravaginal steroids.5 ,12 We prefer to use 2% clindamycin cream intravaginally for 7 days as first line treatment where the AV score is 5 or more, as it is active against staphylococci and streptococci as well as anaerobes. Other treatments that may be of value guided by clinical findings include:

  • Topical oestrogen replacement, such as oestradiol.

  • Intravaginal steroid, for example, Predfoam enema applied intravaginally (off-label use).

  • Oral antibiotics to treat specific pathogenic isolates such Streptococcus pyogenes or Staphylococcus aureus if these are identified.


AV and the more severe presentation of DIV need to be recognised in women presenting with vaginal discharge who have additional symptoms related to vaginal inflammation. A defined scoring system such as the AV score may help identify such women, provoke thought about alternative diagnoses and monitor response to treatment. There are few well conducted controlled treatment trials or guidelines available, but treatments include topical antimicrobials active against aerobic organisms, oestrogen replacement where there is atrophy and topical immunosuppression for severe inflammation. In future, next-generation sequencing will help clinicians better appreciate the diversity of the vaginal microbiome and may eventually replace traditional microscopy and culture-based approaches.


Further reading

View Abstract


  • Twitter Follow Mark Mason at @Mark Mason and A Winter at @teowinterandy

  • Contributors Both authors contributed equally to this manuscript.

  • Competing interests None declared.

  • Provenance and peer review Commissioned; externally peer reviewed.